SI
SI
discoversearch

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor.  We ask that you disable ad blocking while on Silicon Investor in the best interests of our community.  If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.
Technology Stocks : Advanced Micro Devices - Moderated (AMD)
AMD 210.00-2.0%3:59 PM EST

 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  
To: Joe NYC who wrote (33332)3/24/2001 10:39:13 PM
From: fyodor_Read Replies (1) of 275872
 
Joe: I guess it would be a good idea to define what these states are. My guess is that these are not states but "programs" specifying CPU behaviour. If Andreas is right about the granularity of the CPU changes to be down to 1 MHz, there are far more states. If you add to it the changes of voltages, you have even more states.

Andreas and I are, as far as I can tell, in complete agreement on the definition of "what these states are". I certainly leave room for the possibility that I have just misunderstood what I've read of the 6 / 32 states ;). However, I clearly remember some AMD guy talking about how PowerNow! was better than SpeedStep, because PowerNow! had 6 throttle levels (with 32 to be introduced), whereas SpeedStep only had 2. This was quite a while ago, and as I've mentioned, I was unable to find a link to anything like that :(

IMO the changes of just the multiplier would be good enough. Isn't changing FSB also messing with PCI and AGP cards?

That's why I included the FSB+X option.

Basically, there are two ways of getting around the problem you mention:

- Using the memory bus clock for PCI and AGP.

- Using FSB+X for PowerNow! states.

The latter is already used in a FSB = MEMCLOCK+X in some Athlon (KT133) motherboards today.

-fyo
Report TOU ViolationShare This Post
 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext